Home NVIDIA CMP 50HX

NVIDIA CMP 50HX

This is a GPU manufactured with TSMC 12nm process, based on Nvidia Turing architecture and released on Jun 2021. It features 18.6 billion transistors, 3584 CUDA cores and 10GB GDDR6 memory, with 5MB L2 cache, theoretical performance of 11.07TFLOPS, with total power consumption of 250W.

Graphics Card

[Report Issues]
Release Date
Jun 2021
Generation
Mining GPUs
Type
Desktop
Bus Interface
PCIe 3.0 x16

Clock Speeds

[Report Issues]
Base Clock
1350MHz
Boost Clock
1545MHz
Memory Clock
1750MHz

Memory

[Report Issues]
Memory Size
10GB
Memory Type
GDDR6
Memory Bus
320bit
Bandwidth
560.0 GB/s

Render Config

[Report Issues]
SM Count
56
Shading Units
3584
TMUs
192
ROPs
80
Tensor Cores
448
RT Cores
56
L1 Cache
64 KB (per SM)
L2 Cache
5MB

Theoretical Performance

[Report Issues]
Pixel Rate
123.6 GPixel/s
Texture Rate
296.6 GTexel/s
FP16 (half)
22.15 TFLOPS
FP32 (float)
11.07 TFLOPS
FP64 (double)
346.1 GFLOPS

Graphics Processor

[Report Issues]
GPU Name
TU102
GPU Variant
TU102-100-A1
Architecture
Turing
Foundry
TSMC
Process Size
12 nm
Transistors
18,600 million
Die Size
754mm²

Board Design

[Report Issues]
TDP
250W
Suggested PSU
600W
Outputs
No outputs
Power Connectors
2x 8-pin

Graphics Features

[Report Issues]
DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
CUDA
7.5
Shader Model
6.6

Rankings

[Report Issues]
FP32 (float)
11.15 TFLOPS
11.15 TFLOPS
NVIDIA CMP 50HX 10 GB GDDR6
11.07 TFLOPS
11.06 TFLOPS
11.06 TFLOPS
11.06 TFLOPS
10.97 TFLOPS
Blender
1417
NVIDIA CMP 50HX 10 GB GDDR6
1384
1316
1295
NVIDIA CMP 40HX 8 GB GDDR6
1294

Related Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy